Self-defeating behavior patterns constitute a formidable barrier to healthy, adaptive functioning, as well as impairing one's ability to compete effectively and manage one's affairs in a rational, successful fashion. Moreover, insofar as self-defeating behavior runs counter to the rational pursuit of self-interest, it challenges psychology's basic assumptions about human nature. Decades of research have confirmed beyond a doubt that people do engage in systematic patterns of self-defeat, yet these patterns have failed to conform to preliminary theories based on death wishes, desire for punishment, or self-destructive motives. The proposed research will explore two major causes of self-defeating behavior, namely overcommitment arising from self-aggrandizing responses and risk-prone judgment styles elicited by emotional distress. The first part of the proposed work investigates the hypothesis that people sometimes defeat themselves by committing themselves to goals and obligations that exceed their capabilities. Often people can choose their goals and obligations; setting excessively high goals can lead to disaster. The pervasive human tendency to overestimate one's capabilities and strengths increases the likelihood of this costly mistake. In particular, people who have high self-esteem but receive some esteem- threatening news, such as an initial failure or setback, tend to respond by taking an extremely self-aggrandizing pattern that can lead into unrealistic commitments. Goals that exceed one's capabilities will by definition lead to failure; in addition, very high goals may increase the sense of performance pressure which may in turn hamper the person's ability to perform effectively, thus constituting a further obstacle to effective and healthy functioning. This research will also provide a much-needed counterweight to the predominant tendency to view high self- esteem as an unmitigated benefit and as a foundation of healthy adjustment. Both laboratory experiments and first-person accounts of actual experiences will be used in this research. The second part of the proposed research will show how emotional distress alters patterns of judgment and risk appraisal, leading to a tendency to make poor, nonoptimal decisions that increase the likelihood of bad outcomes. More precisely, unpleasant emotional states foster a preference for high-payoff gambles that carry an increased risk of misfortune. In such states, people seem drawn toward activities that offer a chance (however small) of a very positive outcome, presumably because these outcomes might help the person to escape from the bad mood. Unfortunately, people in such a state appear to discount or ignore the likelihood of bad outcomes, and so many of the choices they make will lead to grief, which would perhaps intensify the emotional distress. Emotional suffering may therefore become part of self-perpetuating cycle which would make health, happiness, and adjustment impossible. Again, both laboratory simulations and first-person accounts of actual experiences will be used to do this research.